browser-use
Verified · 14 days agoBrowser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Scrape Booking.com, Amazon reviews, Zillow, Reddit, Google Jobs/News/Flights via Apify's MCP
claude mcp add --transport http crawlkit-scrapers https://mcp.apify.com/?actors=sweet_stickball/booking-scraper,sweet_stickball/amazon-reviews-scraper,sweet_stickball/zillow-scraper,sweet_stickball/reddit-scraper,sweet_stickball/google-jobs-scraper,sweet_stickball/google-news-scraper,sweet_stickball/google-flights-scraper,sweet_stickball/shopify-store-scraper,sweet_stickball/telegram-public-channel-scraper,sweet_stickball/bing-ddg-serp
Exposes Apify-backed scrapers for several travel, commerce, property, social, jobs, and news sites. It suits automation workflows needing structured public-web data, though the broad set of site-specific integrations, zero-star footprint, and lack of a listed package call for careful testing and attention to target-site terms and data-handling rules.
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.
MCP server for browser use access
Playwright Tools for MCP
draw.io diagrams your AI agent draws on a live browser canvas that you can also edit by hand
AI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.